Chromatography Forum

Hello,
I am working on a GC method being developed by another site. The current method calls for nitrogen as the carrier gas at 20 ml/min. The site has switched to helium due to the unavailability of nitrogen in their country. They are running helium at 20 ml / min. Here are the analytical parameters:

-GC: PerkinElmer GC AutoSystem XL with autosampler, data collected with Totalchrom C/S rel. 6.30
-Column: column packed with OV17 3% Chromosorb W-HP 100-120 mesh 1.8 m x 2 mm ID
-Carrier gas: Helium at 20 mL/min
-Column temperature: 210°C
-Injector port temperature: 250°C
-Detector temperature: 250°C

I know that flow rate conversion from nitrogen to helium is 1:1. Does anyone have a formula or know of a mathematical conversion?

Try to download from the Agilent site: GC pressure/flow calculator software and GC Method Translation software at:

http://www.chem.agilent.com/cag/main.html

You can easily then calculate the required nitrogen flow for your method

Best wishes

But it's a packed column so the programs will not know the impedance of the column.

Off the subject but still very interesting to me...

Your foreign country cannot get nitrogen but they can get helium?

Fascinating.

Best regards.

Off the subject but still very interesting to me...
Your foreign country cannot get nitrogen but they can get helium?

Not sure if it applies in this case, but over the past decade here there has been a general move to separate medical gases from industrial use of those gases.

If high purity nitrogen is predominantly used in medical environments, then industrial labs can't access it any more ( danger of cylinder contamination, even with residual pressure valve in-built on cylinder fitting ).

If there isn't enough demand, the cost of high purity gases goes way up. It may be that somebody locally is already using high purity helium, so it's held in stock, and readily available, and cheaper than importing a high purity nitrogen cylinder.

I'm not certain what the original poster wants to do. It's a packed column, flow is set at 20 ml/min, and easily checked by soap film flowmeter. If the separation is OK, what's the problem?.

Please keep having fun,

Bruce Hamilton

It is true that the difference between using nitrogen and helium for most packed column applications is minimal.

Just set your flow for helium at the same flow you set originally for nitrogen. You should notice nothing except perhaps the helium will be slightly faster of a few percent retention time.

best wishes,

Rod

Optimal flow rate of nitrogen in a thin film capillary column is about 3.3 times lower than optimal flow rate of helium in the same column. This ratio can be used as a first approximation for method translation in packed columns. lmb

To continue the topic - how do you correct for the impedence from the column?

Thanks,
Mike

Hello Mike. I do not understand you question. lmb

Mike,

You asked: "- how do you correct for the impedence from the column? "

I think the answer is EXPERIMENTALLY.

Since in theory helium should produce slightly faster retention times, you adjust the flow until you achieve similar retention times that you produce when you use nitrogen as the carrier.

But I suspect that the time change will be less than you think.

best wishes,

Rod

Ok, I was hoping there was a mathematical algorithm to follow that would give me some guidance here. As soon as I get the experimental data I will be sure to post the reply.

My question "impedence of the column" was how much would say packing material / column diameter / column length impede the flow of helium vs. hydrogen.

Our problem is that since the one site is using helium and did not see a peak, and another site using nitrogen was seeing a peak.

Could the site using helium be theoretically not resolving the peak at the flow rate for helium at 20 ml/min since nitrogen theoretically can produce a greater efficiency over a more narrow flow window? What would I need to increase or decrease the helium rate to be to make the conditions experimentally similar?

So do you have any experimental data to share?

Or did the backpressure difference between two different packed columns have more influence than you thought it would?

best wishes,

Rod